Fuchs Illuminates Associations Between Cancer, Lifestyle Habits

Charles S. Fuchs, MD

The idea that by choosing your foods carefully you can improve your resistance to cancer and your survival odds has gained much credibility through the years, largely owing to the work of one man in particular—Charles S. Fuchs, MD, MPH.

Fuchs, the 2017 winner of the Giants of Cancer Care^® award in Gastrointestinal Cancer, began using population studies more than 2 decades ago to clarify the roles that diet, exercise, and other very basic lifestyle choices play in both the prevention and treatment of gastrointestinal cancers. In doing so, Fuchs and his collaborators have demonstrated strong ties between cancer and an ever-growing list of items: vitamin D, aspirin, red meat, exercise, obesity, and more.

“I became interested, right at the very beginning of my professional life, in identifying risk factors for gastrointestinal cancers, and it occurred to me that I might spot them by analyzing data from a project that had been going on since long before I arrived on the scene—the Harvard cohort studies,” says Fuchs.

His academic work, which has seen him lead major drug trials and perform many other forms of cancer research, has resulted in more than 600 published papers. His administrative work, which saw him lead the Gastrointestinal Malignancies Program at Dana-Farber Cancer Center for nearly a decade, inspired Yale University to name him director of its rapidly growing cancer center. Yet Fuchs remains most well-known for the epidemiological work that started with cohort studies from Harvard’s Pooling Project of Prospective Studies of Diet and Cancer.

Early Days to Harvard

“There were, and are, 2 huge cohorts, female nurses and male health professionals, with more than 200,000 combined participants, all providing a huge amount of health-related personal information over decades of time. Such large populations were clearly going to develop a large number of gastrointestinal cancers over the years, and it was my hope that the numbers would be large enough and the patient information comprehensive enough to test hypotheses and identify risk factors with a high degree of certainty,” says Fuchs, whose idea proved sound enough to furnish years of new discoveries.Fuchs was born in the Bronx and grew up about 8 miles farther north, in New Rochelle, New York. His father was in the moving and storage business, and his mother was a homemaker.

He describes himself, in retrospect, as a good student—good enough to get into the University of Pennsylvania—but not exceptional. He developed an interest in medicine during his high school years, when he volunteered at a small local hospital that gave him a wide variety of tasks. The young Fuchs registered patients, wheeled people around, did simple lab work, assisted radiologists—whatever was needed on any given day. The experience gave him a broad overview of what different doctors did and his first inclination toward cancer care.

Fuchs committed to oncology during his years at Harvard Medical School and started to specialize in gastrointestinal cancers while doing his residency at Brigham and Women’s Hospital of Boston and studying for a master’s degree at the Harvard School of Public Health. The decision stemmed, in part, from the realization that the statistical skills he learned at the School of Public Health lent themselves to the analysis of such a common type of cancer and, in part, from the fact that it was the chosen specialty of Robert J. Mayer, MD, one of his mentors, who would later be named a 2015 Giants of Cancer Care® winner.

At that time, Mayer had already made a name for himself, not only as a great researcher but also as a great mentor, the sort who would give young colleagues the opportunity to do interesting and important work and offer the support they’d need to navigate the complexities of grant requests, research oversight committees, and the peer review process. Not that Fuchs needed all that much support.

“I vividly recall that we collaborated to write a review paper on gastric cancer for the New England Journal of Medicine [NEJM], just when Charlie was starting out. He submitted his manuscript to me, and I got out my red pen, because he was at that age where I knew he’d need a huge amount of editing. Then I read through the whole thing without making a mark,” Mayer says. “His ability to take a huge amount of information and present it all clearly and logically was amazing right from the very start. It shows in everything he does, but it is particularly important in epidemiological work.”

The Harvard health cohorts are, for those who enjoy data analysis, an almost inexhaustible mine of medical insight. Because they collect detailed information about huge numbers of people decades before most of the participants develop any serious health problems, they are free from selection bias and thus a unique tool for identifying early risk factors for virtually any serious disease. Researchers can look for links between any metric or behavior the studies track and any common health outcome.

By testing hypothesis after hypothesis, Fuchs and his various collaborators demonstrated previously unsuspected links between many lifestyle factors and cancer incidence. An analysis of 88,410 middle-aged nurses from 1980 to 1994, for example, found that those who ate beef, pork, or lamb daily were more than twice as likely to get non-Hodgkin lymphoma as those who ate such meats less than once a week. Another analysis associated working at least 3 nights a week for at least 15 years with an increased risk of colorectal cancer. A third study found a 30% reduction in colorectal cancer incidence among women who took more than 14 aspirin a week for more than a decade.

Fuchs and his collaborators kept coming back to the data and mining it for new insights. The associations they found provided the material for dozens of papers, and that early work established his reputation in more ways than one.

“It was apparent very early on that Charlie was a great collaborator,” says Edward Giovannucci, MD, MPH, ScD, a professor of nutrition and epidemiology at the Harvard T.H. Chan School of Public Health (renamed in 2014). “He always does his share and gets it done on time, no matter how much else he has on his plate. I’ve been working with him for more than 20 years, and we’ve never had a conflict that wasn’t resolved in minutes. That’s very unusual in any academic collaboration that lasts more than a few months, let alone a few decades....People hear about that, and they want to work with him.”

Cancer Incidence, Not Prevention

Of course, not every hypothesis panned out. In a study of 88,757 women, Fuchs and his collaborators found that dietary fiber intake offered no protection against colorectal cancer, and fruit and vegetable consumption throughout the 1980s was not significantly associated with colon cancer rates as of 1996. But even negative findings such as these provided important enough information to garner several hundred citations apiece.

At first glance, the importance of such discoveries may seemingly lie in cancer prevention, but Fuchs doubts they will inspire the behavioral changes that will reduce incidence. “We already knew that obesity is dangerous,” Fuchs says. “The discovery that it increases the risk of gastrointestinal cancers isn’t going to make fitness nuts out of millions of Americans who ignored the risk of diabetes and cardiovascular events.

“The real value in the discovery that something unexpectedly increases cancer incidence is that you can then go to the laboratory, trace the pathway, and learn something new about the cancer. What’s more, you can also see whether the same factor affects outcomes among people who already have the cancer.”

Even the largest general cohorts never accumulate enough patients with gastrointestinal cancer to determine whether lifestyle factors affect outcomes after diagnosis, so Fuchs and his colleagues at Dana-Farber began a patient cohort in 1999. Immediately after their tumor masses were resected, the patients were enrolled in the study and the same information was tracked as in the larger population cohorts.

Subsequent analysis found, contrary to the expectations of almost everyone, that seemingly minor lifestyle factors were associated with very significant changes in cancer recurrence and mortality. Mortality rates were 25% to 30% lower among patients who took at least 3 adult aspirin every week than they were among similar patients who did not, 35% lower among patients who maintained healthy vitamin D levels, and 45% lower among patients who exercised regularly.

It is tempting to perform a bit of mental addition and conclude that patients who began a regimen of both exercise and supplements right after surgery could nearly eliminate the risk of cancer death, but Fuchs warns against such assumptions. Cohort studies find associations, not causes. It is possible that some underlying difference between aspirin users and nonusers— something besides the aspirin itself—affects outcomes. Even if you assume the studied behaviors are changing the outcomes, Fuchs says, the benefits would likely overlap rather than be fully additive.

“Still,” he adds, “the number of associations and the strength of those associations argue that lifestyle choices made after the initial diagnosis will eventually prove to be major factors in patient outcomes. Making the right choices regarding diet and exercise still won’t be easy for many patients, but a cancer diagnosis is a major motivator.”

Research done mostly by others suggests that lifestyle may also play a large role in determining outcomes among patients with other tumor types, especially breast cancer, but any such link would be particularly important for patients with gastrointestinal cancers. Even though they collectively account for nearly a quarter of all cancer diagnoses, such tumors have seen fewer new treatments than many other tumor types.

Indeed, until 2014, when the FDA approved a novel angiogenesis inhibitor called ramucirumab (Cyramza) on the strength of a trial led by Fuchs, the only approved treatments for gastric cancer (the fifth most common of all cancers) were chemotherapies originally developed for other tumor types. Even chemotherapy treatment for gastric cancer is relatively new: It wasn’t until 2001, when targeted therapies were already transforming the treatment of many tumors, that the FDA first approved a chemotherapy for adjuvant use in patients with gastric cancer.

Balancing the Professional and Personal

Fuchs hoped to improve the treatment of gastric cancer further still and, in September 2017, his wish was granted when the immune checkpoint inhibitor pembrolizumab (Keytruda) was approved for the treatment of recurrent locally advanced or metastatic gastric cancers, thereby making it the first immunotherapy approved in this tumor type. He is hopeful that advances in immunotherapy will continue to influence patients with gastric cancers.Fuchs has balanced this research with a heavy administrative load for years. He was only the second specialist in Dana-Farber’s gastrointestinal cancer division when he joined Mayer in the early ’90s, but he worked to expand the division, first as Mayer’s deputy and then, after Mayer’s 2008 retirement, as division head. When Fuchs left for Yale early in 2017, Dana- Farber had 28 gastrointestinal specialists.

The move from head of a division to head of an entire cancer center will necessarily entail even more administrative work, but Fuchs still plans to make time not only for research but also for patient care. “I work long days, but I’m passionate about what I do, and I think I can be effective as a researcher, a clinician, and an administrator if I am strategic about how I use my time,” he says, adding that significant achievement in medical research requires hard work—but not so much hard work that it precludes having a normal life.

“My advice to anyone reading this who has an interest in doing research is to pursue it,” Fuchs says. “It is hard to describe the satisfaction that comes with such work, and it does not require the sacrifice of normal life pleasures. I have been very happily married for more than 25 years. I have coached my kids in Little League and soccer. I go on vacation. It’s very possible to do this work and have a life.”

The Harvard health cohorts were large enough to weigh the costs and benefits of certain behaviors, at least when it came to very common medical problems. For example, Fuchs was the lead author of a 1995 paper that assessed the impact of drinking among Nurses’ Health Study participants. The analysis, which appeared in NEJM, found that light to moderate drinking was associated with a lower relative risk of death than abstinence, whereas heavy drinking was associated with a higher risk of death. However, sub-analysis found that the benefits of drinking came mainly from reduced cardiovascular events, while the drawbacks came mainly from increased rates of cirrhosis and breast cancer. Therefore, the study concluded, women who otherwise faced a high risk of breast cancer and a low risk of cardiovascular disease faced a higher risk of death from any degree of drinking. It was a remarkable demonstration of how finely the data could be parsed, and it has been cited 763 times to date.